Re: [patch, libgfortran] Part 2: PR105456 Child I/O does not propage iostat

[Jerry D <jvdelisle2@gmail.com>]

On 2/29/24 12:56 PM, Steve Kargl wrote:
> On Thu, Feb 29, 2024 at 10:28:19AM -0800, Jerry D wrote:
>> On 2/29/24 10:13 AM, Steve Kargl wrote:
>>> On Thu, Feb 29, 2024 at 09:36:43AM -0800, Jerry D wrote:
>>>> On 2/29/24 1:47 AM, Bernhard Reutner-Fischer wrote:
>>>>
>>>>> And, just for my own education, the length limitation of iomsg to 255
>>>>> chars is not backed by the standard AFAICS, right? It's just our
>>>>> STRERR_MAXSZ?
>>>>
>>>> Yes, its what we have had for a long lone time. Once you throw an error
>>>> things get very processor dependent. I found MSGLEN set to 100 and IOMSG_len
>>>> to 256. Nothing magic about it.
>>>>
>>>
>>> There is no restriction on the length for the iomsg-variable
>>> that receives the generated error message.  In fact, if the
>>> iomsg-variable has a deferred-length type parameter, then
>>> (re)-allocation to the exact length is expected.
>>>
>>>     F2023
>>>
>>>     12.11.6 IOMSG= specifier
>>>
>>>     If an error, end-of-file, or end-of-record condition occurs during
>>>     execution of an input/output statement, iomsg-variable is assigned
>>>     an explanatory message, as if by intrinsic assignment. If no such
>>>     condition occurs, the definition status and value of iomsg-variable
>>>     are unchanged.
>>> character(len=23) emsg
>>> read(fd,*,iomsg=emsg)
>>>
>>> Here, the generated iomsg is either truncated to a length of 23
>>> or padded with blanks to a length of 23.
>>>
>>> character(len=:), allocatable :: emsg
>>> read(fd,*,iomsg=emsg)
>>>
>>> Here, emsg should have the length of whatever error message was
>>> generated.
>>> HTH
>>>
>>
>> Well, currently, if someone uses a larger string than 256 we are going to
>> chop it off.
>>
>> Do we want to process this differently now?
>>
> 
> If I look at the interfaces for UDTIO in F2023 (pp. 254-255),
> the declaration for iomsg is
> 
>    CHARACTER (LEN=*), INTENT(INOUT) :: iomsg
> 
> F2023 (p. 62) has
>    An asterisk as a type-param-value specifies that a length type
>    parameter is an assumed type parameter.  It is used for a dummy
>    argument to assume the type parameter value from the effective
>    argument, ...
> 
> So, in theory, one should be able to get the required length
> from length from the argument.
> 
>    CHARACTER(LEN=23) str
>    WRITE(6,'(DT)',iomsg=str) derived-type-entity
> 
> If the subroutine supplied by the user internally creates
> an error message with 300 characters, then from the above I
> would think that it will be truncated to 23 characters.
> OTOH, if the user is expecting the full 300 characters with
> 
>    CHARACTER(LEN=300) str
>    WRITE(6,'(DT)',iomsg=str) derived-type-entity
> 
> then truncating internally to 256 seems wrong.
> 
> Now, that I looked at the interface more closely, the declaration
> 
>    CHARACTER (LEN=*), INTENT(INOUT) :: iomsg
> 
> seems to block the use of an unallocated deferred-length 'str'
> and (re)-allocation does not occur.
> 
>   
> 
Without addressing the length discussions above, I did find an existing 
function in libgfortran to trim the spaces at the end (string_len_trim). 
  I am using it as follows:

		if ((dtp->u.p.child_saved_iostat != 0) &&
		    !(dtp->common.flags & IOPARM_HAS_IOMSG) &&
		    !(dtp->common.flags & IOPARM_HAS_IOSTAT))
		  {
		    char message[IOMSG_LEN];
		    child_iomsg_len = string_len_trim (IOMSG_LEN, child_iomsg) + 1;
		    snprintf (message, child_iomsg_len, child_iomsg);
		    generate_error (&dtp->common, dtp->u.p.child_saved_iostat,
				    message);
		    goto nml_err_ret;
		  }

I will study the len questions further.

Jerry

PS I wish my mail client would not wrap text on me, working on that issue.